Sort and specify the quartile rule
Use at least two real observations on the same numerical scale. Sort ascending and retain repeats. Split an even count into equal halves; for an odd count exclude the overall median from both halves. Q1 is the lower-half median and Q3 the upper-half median. This explicitly chosen rule is ISB variant 2, not a universal definition across software.[1]
An odd-length list has its middle observation as median; an even-length list uses the average of its two middle observations. For n=2 each half contains one value. Sorting and splitting give the ordered summary minimum≤Q1≤median≤Q3≤maximum. Distances depend on measured values rather than drawn pixels.[1]
IQR, two whisker rules and outliers
The IQR and fences have the data’s unit; 1.5 is dimensionless.[2]
A basic min–max boxplot extends whiskers to the minimum and maximum, as in the Bavarian school material. The modified Tukey-style plot added here extends them only to the outermost actual observations on or inside the 1.5-IQR fences. Values strictly outside are individually flagged. The quartile algorithm remains a separate convention to report.[1][2]
A box gives a coarse picture of central location and quartile spread. Ties and the chosen short-list convention mean it does not always contain exactly half the observations. An outlier flag indicates something unusual, not proof of a measurement error. No normality assumption is needed for this description.[1]
Evaluate odd and even lists completely
Nine values with one flagged maximum
- Given
- Sorted data: 1,2,4,6,7,8,10,12,32.
- Find
- Find quartiles, IQR, fences and modified whiskers.
For odd n exclude the overall middle observation from both halves.
Take the medians of the halves.
Calculate spread and the two fences.
Use actual observations within the fences as whisker endpoints.
Result: Q1=3, median 7, Q3=11, IQR 8; fences −9 and 23. 32 is flagged. Modified whiskers end at 1 and 12.
Check: Eight observations are within the fences and one is above. A simple min–max boxplot would extend to 32; the modified version shows 32 separately. No whisker ends at the calculated fence 23.
An even count with repeated observations
- Given
- Sorted data: 2,3,3,5,7,8,10,14.
- Find
- Find the five-number summary and check for outliers.
Retain all eight observations and split into two halves of four.
Average the middle pair of each half.
Compute the IQR and fences.
Compare the minimum and maximum with the fences.
Result: Five-number summary(2,3,6,9,14). No value is outside the fences, so both whisker conventions end at 2 and 14.
Check: The median 6 need not be observed. The two occurrences of 3 remain two data points. The ordering min≤Q1≤M≤Q3≤max and IQR≥0 both hold.
Box, observed whiskers and a separate outlier
Free with a Quanta account. Your download stays ready after registration.
Recall the method later with the formula’s flashcards.
10 flashcards · English materials
Already have an account? Sign in for flashcardsAvailable from Essential. A ready-made formula set, no AI credits required. Article exercises are not all included.
List limits and reproducible results
The list calculator uses two to one hundred finite real observations without weights, the stated median-of-halves rule, and strict-outside comparisons at the fences. It can determine Q1, median, Q3, IQR, fences and the count of flagged observations. Grouped frequency tables, censored data, automatic error cleaning and all software quantile types are outside this model. Each observation additionally requires |x|≤1012 in the calculator.
Five solved exercises
Calculate · Foundation
Find Q1, median, Q3 and IQR for 1,4,6,8,10,15.
Show hint
Split the sorted list into two groups of three.
Learning goal: Apply medians to an even list and its halves.
Show solution and calculation
L=(1,4,6),U=(8,10,15).
Thus Q1=4,M=(6+8)/2=7,Q3=10,IQR=6.
Fences are −5 and 19, with no observations outside.
Check: 4≤7≤10 and 10−4=6.
Rearrange · Application
Q1=8 and the upper 1.5-IQR fence is 26. Find Q3 and IQR.
Show hint
Substitute IQR=Q3−8 into the fence equation.
Learning goal: Rearrange a fence relationship and recognise missing information.
Show solution and calculation
26=Q3+1.5(Q3−8)=2.5Q3−12.
Thus Q3=38/2.5=15.2 and IQR=7.2.
Check: 15.2+1.5·7.2=26, and Q3≥Q1. Without a data list, neither the median nor whiskers are uniquely determined.
Interpret · Application
Compare A=(2,4,5,6,8) and B=(0,4,5,6,10) using median and IQR.
Show hint
For five observations exclude the middle value from both halves.
Learning goal: Distinguish equal median location from different quartile spread.
Show solution and calculation
Both medians are 5.
For A,Q1=3,Q3=7,IQR=4; for B,Q1=2,Q3=8,IQR=6.
Under this convention B has a wider box at the same median.
Check: the full lists show a location-and-spread description, not evidence about causes or statistical significance.
Error analysis · Application
For 0,4,4,6,8,8,20 the upper fence is 14. A diagram puts the whisker there. Correct it.
Show hint
A modified whisker endpoint is an observed value.
Learning goal: Distinguish fences, observed endpoints and boxplot conventions.
Show solution and calculation
Q1=4,M=6,Q3=8,IQR=4; fences are −2 and 14. 20 is outside, and the largest observation inside is 8.
The upper whisker therefore ends at 8, with 20 shown separately.
Check: 14 is absent from the list. A simple min–max boxplot would instead end at 20; state which variant is intended.
Transfer · Application
Waiting times in minutes: 4,5,5,6,6,7,7,8,30. Summarise robustly and assess 30.
Show hint
An outlier flag does not prove a measurement error.
Learning goal: Include units and data checking in a statistical summary.
Show solution and calculation
M=6min,Q1=5min,Q3=7.5min,IQR=2.5min.
Fences are 1.25min and 11.25min, so 30min is flagged; whiskers end at 4min and 8min.
Check: 30>11.25. Keep the observation documented and investigate it rather than deleting it solely because of the rule. The median describes central waiting time without implying that every wait is short.
Check your understanding with the formula’s reference test.
8 test questions · English materials
Already have an account? Sign in for the testAvailable from Essential. A ready-made formula set, no AI credits required. Article exercises are not all included.
Compare location and other spread measures
Mean and median explain dataset location and their different responses to extreme values. Sample variance instead uses squared deviations from the mean, addressing a different spread question from the interquartile range.
Sources for this explanation
Quanta · Updated November 5, 2026
- ISB, Descriptive Statistics: Summary Measures and Boxplots, Quartile Convention 2 and Whisker Conventions (German-language source)
- NIST/SEMATECH e-Handbook,1.3.3.7 Box Plot: basic plot and fences
- LehrplanPLUS Bayern, Gymnasium Mathematik7, M7.4 Kenngrößen von Daten
Editorial responsibility: Quanta Study editorial team
Review scope
This explanation, 2 worked examples and 5 exercises are available in German and English. Structure, source references and mathematical notation undergo automated checks. Calculations, model assumptions and both language versions receive a separate AI review.
An independent, named human subject review has not yet been documented. Our exercise numbers are worked examples, rather than measurements published by the cited sources.
Editorial responsibility, sources and methodsChange history
- Authored an independent DE/EN school article with a justification, two original examples, five solved exercises and a native diagram.
Cite this reference page
Quanta. Quartiles, interquartile range and boxplots. Updated November 5, 2026. https://quanta-study.de/en/formula-sheet/quartile-boxplot
When citing a specific scientific statement, also cite the specialist source linked to that statement.